How to Reduce Minor Stops

Minor stops are brief unplanned stoppages an operator clears without maintenance, a jam, a misfeed, a tripped sensor, a part that hangs up, usually lasting seconds to a couple of minutes. You reduce them by logging the pattern, not the incident: count each stop by cause and location, then fix the feed, alignment, or sensor problem at its source rather than resetting the same fault all shift.

Minor stops are the sneakiest loss on the floor because each one is trivial and the pile is enormous. Nobody writes up a 40-second jam, so it never reaches the downtime log, and the crew stops seeing them as a problem at all, clearing them becomes part of the job. Meanwhile they quietly drain the OEE Performance factor and wreck flow. This guide shows where minor stops hide, how to surface the pattern behind chronic ones, and how to eliminate the causes instead of clearing the symptoms.

What are minor stops in manufacturing?

Minor stops, also called micro-stops or short stops, are short, self-cleared interruptions to production, typically under two to five minutes, that the operator resolves on the spot without calling maintenance. Common forms are product jams, misfeeds, a photo-eye or proximity sensor false-tripping, a reject-gate hang-up, a label misapply, or a part that needs nudging into place. They differ from breakdowns in three ways: they are brief, they are frequent, and they are cleared by the operator, not repaired by a technician.

In the six big losses framework, this is the loss called "idling and minor stoppages," and it is distinct from breakdowns. That distinction matters because it changes where the loss lands in OEE and, therefore, what fixes it. A breakdown is a discrete, loggable event that hits Availability. A cloud of 90-second stops behaves like a speed loss and hits Performance, a different factor, a different countermeasure.

Why don't minor stops show up in your downtime log?

Because they fall below the threshold anyone bothers to record. A downtime log built on operator memory or a two-minute cutoff captures the breakdown and drops the jam. The stop still happened, so the machine still made fewer units than its ideal speed implies, but instead of appearing as downtime, the lost minutes reappear as a lower Performance factor, where they look like the line simply running slow.

This is the crux: minor stops migrate from Availability, where they belong, into Performance, where they hide. On the Performance factor they are indistinguishable from a machine set too slow, so people chase phantom speed problems while the real cause is dozens of tiny stops. Surfacing them is the whole game. Automatic counting from the machine, every stop event timestamped, however short, drags the loss back into the light where the crew can act on it. Until you can see the stops, you are tuning a symptom.

How do you find the pattern behind chronic minor stops?

Stop clearing them one at a time and start counting them by cause and place. A chronic minor stop is a pattern, and patterns only appear when you tally. Here is a sequence that turns a shrug into a work order:

1. Count stops, don't time them. For minor stops, frequency is the signal, not duration. Get an automatic stop count off the machine or run a manual tally for a few shifts.
2. Tag each stop with cause and location. A short reason code and the station: "infeed jam, station 3," "reject-gate hang, sealer." Location plus cause is what points at a root, not just a symptom.
3. Pareto the tags. Sort causes by count. Chronic minor stops obey the 80/20 rule hard, usually two or three causes at two or three stations produce most of the stops. Build the Pareto chart and the target picks itself.
4. Watch the top cause happen. Go stand at the station and observe ten of them. The tally tells you where; your eyes tell you why. Is it every unit or every tenth? Tied to a product changeover, a worn guide, a specific SKU?
5. Ask why it recurs, not why it stopped. Run a quick 5 Whys on the pattern. "The sensor false-trips" → "it's mounted where mist collects" → "no shield" gets you to a fix; "cleared the jam" gets you to the next jam.
6. Fix the source and keep counting. Change the guide, shield the sensor, adjust the feed, then watch the tally for that cause. If the count drops, you fixed a root; if it doesn't, you fixed a symptom.

What actually causes most minor stops?

Most chronic minor stops trace to one of four families: feeding, alignment, sensing, or product variation. Naming the family is half the fix, because each family has its own standard countermeasure.

Feed problems are the most common and the most fixable: a worn infeed guide or a feed rate set too aggressive produces jams every few minutes that everyone has stopped noticing. Alignment issues creep in after changeovers set "by feel" instead of to a stop or a mark. Sensor faults are maddening because the machine is fine, a photo-eye collecting mist or a proximity sensor mounted a hair too far trips on nothing. Material variation is the one operators can't fix at the line; it shows up as a cause that spikes on a single SKU or a single supplier's lot and points upstream to specification or supplier work.

How do you fix minor stops at the source?

Fix the condition that produces the stop, then make the stop impossible or self-evident. Clearing a jam restores the minute you just lost; changing the guide that caused the jam restores every minute after it. The order is: eliminate, then error-proof, then standardize.

Eliminating means correcting the physical cause, replace the worn guide, reset the feed to its true rate, shield the sensor, set the changeover to a positive stop instead of by eye. Error-proofing means adding a poka-yoke so the fault can't recur: a fixture that only accepts the part one way, a guide that can't be set wrong. Standardizing means capturing the good condition in standard work so the next changeover reproduces it. Layer an andon signal on top for the stops you can't yet eliminate, so a recurring one raises a flag instead of dissolving into the shift. And where material variation is the cause, the fix lives upstream in specification and supplier work, not at the station.

One more discipline: keep counting after each fix. Chronic minor stops are only "solved" when the tally for that cause stays down over several shifts. A stop count that falls and holds is proof; a cleared jam is not. This is exactly the kind of loss that a live picture off the machine turns from invisible to obvious, the way Harmony surfaces micro-stops from machine signals rather than waiting for someone to write them down (see the platform).

How much OEE do minor stops cost?

More than most plants realize, because the loss is spread thin and hidden inside the Performance factor rather than tallied anywhere.

The framework worth anchoring on is Nakajima's six big losses, in which "idling and minor stoppages" is a category in its own right, popularized in Introduction to TPM (Productivity Press, 1988). The often-quoted 95% "world-class" Performance figure is a reference point, not an audited industry number, and minor stops are usually the largest thing standing between a real line and that mark. For macro perspective, the Federal Reserve's G.17 capacity-utilization release is a reminder that plants run well below theoretical maximums by any method; a chunk of that gap, on many lines, is stops too short to log.

Minor stops are beatable, but only after you can see them. Count by cause, Pareto the causes, fix the feed, alignment, sensor, or material at its source, and confirm the tally stays down. Track the recovery in your manufacturing KPIs alongside a clean downtime picture feed clean inputs into the OEE calculator and see how one plant pulled its hidden losses into view in the CLS case study.